Spaces:
Running
on
T4
Running
on
T4
gabriel chua
commited on
upgrade to llama3.3 70B
Browse files- README.md +2 -2
- constants.py +2 -2
README.md
CHANGED
@@ -18,7 +18,7 @@ short_description: Personalised Podcasts For All - Available in 13 Languages
|
|
18 |
This project is inspired by the NotebookLM tool, and implements it with open-source LLMs and text-to-speech models. This tool processes the content of a PDF, generates a natural dialogue suitable for an audio podcast, and outputs it as an MP3 file.
|
19 |
|
20 |
Built with:
|
21 |
-
- [Llama 3.
|
22 |
- [MeloTTS π](https://huggingface.co/myshell-ai/MeloTTS-English)
|
23 |
- [Bark πΆ](https://huggingface.co/suno/bark)
|
24 |
- [Jina Reader π](https://jina.ai/reader/)
|
@@ -53,7 +53,7 @@ To set up the project, follow these steps:
|
|
53 |
## Usage
|
54 |
|
55 |
1. **Set up API Key(s):**
|
56 |
-
For this project, I am using LLama 3.
|
57 |
|
58 |
2. **Run the application:**
|
59 |
```bash
|
|
|
18 |
This project is inspired by the NotebookLM tool, and implements it with open-source LLMs and text-to-speech models. This tool processes the content of a PDF, generates a natural dialogue suitable for an audio podcast, and outputs it as an MP3 file.
|
19 |
|
20 |
Built with:
|
21 |
+
- [Llama 3.3 70B π¦](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) via [Fireworks AI π](https://fireworks.ai/) and [Instructor π](https://github.com/instructor-ai/instructor)
|
22 |
- [MeloTTS π](https://huggingface.co/myshell-ai/MeloTTS-English)
|
23 |
- [Bark πΆ](https://huggingface.co/suno/bark)
|
24 |
- [Jina Reader π](https://jina.ai/reader/)
|
|
|
53 |
## Usage
|
54 |
|
55 |
1. **Set up API Key(s):**
|
56 |
+
For this project, I am using LLama 3.3 70B hosted on Fireworks API as its JSON Mode supports passing a pydantic object. So, please set the API key as the `FIREWORKS_API_KEY` environment variable
|
57 |
|
58 |
2. **Run the application:**
|
59 |
```bash
|
constants.py
CHANGED
@@ -24,7 +24,7 @@ ERROR_MESSAGE_TOO_LONG = "The total content is too long. Please ensure the combi
|
|
24 |
# Fireworks API-related constants
|
25 |
FIREWORKS_API_KEY = os.getenv("FIREWORKS_API_KEY")
|
26 |
FIREWORKS_MAX_TOKENS = 16_384
|
27 |
-
FIREWORKS_MODEL_ID = "accounts/fireworks/models/llama-
|
28 |
FIREWORKS_TEMPERATURE = 0.1
|
29 |
|
30 |
# MeloTTS
|
@@ -78,7 +78,7 @@ UI_DESCRIPTION = """
|
|
78 |
Generate Podcasts from PDFs using open-source AI.
|
79 |
|
80 |
Built with:
|
81 |
-
- [Llama 3.
|
82 |
- [MeloTTS π](https://huggingface.co/myshell-ai/MeloTTS-English)
|
83 |
- [Bark πΆ](https://huggingface.co/suno/bark)
|
84 |
- [Jina Reader π](https://jina.ai/reader/)
|
|
|
24 |
# Fireworks API-related constants
|
25 |
FIREWORKS_API_KEY = os.getenv("FIREWORKS_API_KEY")
|
26 |
FIREWORKS_MAX_TOKENS = 16_384
|
27 |
+
FIREWORKS_MODEL_ID = "accounts/fireworks/models/llama-v3p3-70b-instruct"
|
28 |
FIREWORKS_TEMPERATURE = 0.1
|
29 |
|
30 |
# MeloTTS
|
|
|
78 |
Generate Podcasts from PDFs using open-source AI.
|
79 |
|
80 |
Built with:
|
81 |
+
- [Llama 3.3 70B π¦](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) via [Fireworks AI π](https://fireworks.ai/) and [Instructor π](https://github.com/instructor-ai/instructor)
|
82 |
- [MeloTTS π](https://huggingface.co/myshell-ai/MeloTTS-English)
|
83 |
- [Bark πΆ](https://huggingface.co/suno/bark)
|
84 |
- [Jina Reader π](https://jina.ai/reader/)
|